Theses and Dissertations from UMD
Permanent URI for this communityhttp://hdl.handle.net/1903/2
New submissions to the thesis/dissertation collections are added automatically as they are received from the Graduate School. Currently, the Graduate School deposits all theses and dissertations from a given semester after the official graduation date. This means that there may be up to a 4 month delay in the appearance of a give thesis/dissertation in DRUM
More information is available at Theses and Dissertations at University of Maryland Libraries.
Browse
4 results
Search Results
Item Metareasoning Strategies to Correct Navigation Failures of Autonomous Ground Robots(2024) Molnar, Sidney Leigh; Herrmann, Jeffrey; Mechanical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Due to the complexity of autonomous systems, theoretically perfect path planning algorithms sometimes fail due to emergent behaviors that arise when interacting with different perception, mapping and goal planning subprocesses. These failures prevent mission success, especially in complex environments that have not previously been explored by the robot. To overcome these failures, many researchers have sought to develop parameter learning methods to improve either mission success or path planning convergence. Metareasoning, which can be simply described as “thinking about thinking,” offers another possible solution for mitigating these planning failures. This project offers a novel metareasoning approach that uses different methods of monitoring and control to detect and overcome path planning irregularities that contribute to path planning failures. All methods for the approaches were implemented as a part of the ARL ground autonomy stack which uses both global and local path planning ROS nodes. The proposed monitoring methods include listening to messages published to the system by the planning algorithms themselves, evaluating for the environmental context that the robot is in, the expected progress methods which use the robot’s movement capabilities to evaluate for progress that has been made from a milestone checkpoint, and the fixed radius methods which use user-selected parameters based on mission objectives to evaluate for the progress that has been made from a milestone checkpoint. The proposed control policies are the metric-based sequential policies which use benchmark robot performance metrics to select the order in which the planner combinations are to be launched, the context-based pairs policies which evaluate what happens when switching between only two planner combinations, and the restart policy which simply relaunches a new instance of the same planner combination. The study evaluated which monitoring and control policies, when paired, contributed to improved navigation performance and which policies contributed to degraded navigation performance by evaluating how close the robot was able to get to the final mission goal. Although specific methods were evaluated, the contributions of the project extend beyond the results by offering both a template for metareasoning approaches with regard to navigation as well as replicable algorithms that may be applied to any autonomous ground robot system. Additionally, this thesis presents ideas for additional research in order to determine under which conditions metareasoning will improve navigation.Item Metareasoning Approaches to Thermal Management During Image Processing(2022) Dawson, Michael Kenneth; Herrmann, Jeffrey W; Mechanical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Resource-constrained electronic systems are present in many semi- and fully-autonomous systems and are tasked with computationally heavy tasks such as image processing. Without sufficient cooling, these tasks often increase device temperature up to a predetermined maximum, beyond which the task is slowed by the device firmware to maintain the maximum. This is done to avoid decreased processor lifespan due to thermal fatigue or catastrophic processor failure due to thermal overstress. This thesis describes a study that evaluated how well metareasoning can manage the central processing unit (CPU) temperature during image processing (object detection and classification) on two devices: a Raspberry Pi 4B and an NVIDIA Jetson Nano Developer Kit. Three policies that employ metareasoning were developed; one which maintains a constant image throughput, one which maintains a constant expected detection precision, and a third that trades between throughput and precision losses based on a user-defined parameter. All policies used the EfficientDet series of object detectors. Depending on the policy, these networks were either switched between, delayed, or both. This thesis also considered cases that used the system's built-in throttling policy to control the temperature. A policy was also created via reinforcement learning. The policy was able to adjust the detection precision and program throughput based on a set of states corresponding to the possible temperatures, neural networks, and processing delays. All three designed metareasoning policies were able to stabilize the device temperature without relying on thermal throttling. Additionally, the policy created through reinforcement learning was able to successfully stabilize the device temperature, though less consistently. These results suggest that a metareasoning-based approach to thermal management in image processing is able to provide a platform-agnostic and programmatic way to comply with constant or variable temperature constraints.Item A Method For Improving Decentralized Task Allocation For Multiagent Systems in Low-Communication Environments.(2021) Akoroda, Oghenetekevwe; Herrmann, Jeffrey W; Systems Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Communication is an important aspect of task allocation, but it has a cost and low communication restricts the information exchange needed for task allocation. As a result, a lot of decentralized task allocation algorithms perform worse as communication worsens. The contribution of this thesis is a method to improve the performance of a task allocation algorithm in low-communication environments and reduce the cost of communication by restricting communication. This method, applied to the Consensus Based Auction Algorithm (CBAA), determines when an agent should communicate and estimates the information that will be received from other agents. This method is compared to other decentralized task allocation algorithms at different levels of communication in a ship protection scenario. Results show that this method when applied to CBAA performs comparably to CBAA while reducing communication.Item DECENTRALIZED MULTIAGENT METAREASONING APPLICATIONS IN TASK ALLOCATION AND PATH FINDING(2021) Langlois, Samuel; Herrmann, Jeffrey W.; Systems Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Decentralized task allocation and path finding are two problems for multiagent systems where no single fixed algorithm provides the best solution in all environments. Past research has considered metareasoning approaches to these problems that take in map, multiagent system, or communication information. None of these papers address the application of metareasoning about individual agent state features which could decrease communication and increase performance for decentralized systems. This thesis presents the application of a meta-level policy that is conducted offline using supervised learning through extreme gradient boosting. The multiagent system used here operates under full communication, and the system uses an independent multiagent metareasoning structure. This thesis describes research that developed and evaluated metareasoning approaches for the multiagent task allocation problem and the multiagent path finding problem. For task allocation, the metareasoning policy determines when to run a task allocation algorithm. For multiagent path finding, the metareasoning policy determines which algorithm an agent should use. The results of this comparative research suggest that this metareasoning approach can reduce communication and computational overhead without sacrificing performance.